Malondialdehyde (MDA) and structurally related dicarbonyls and acroleins are natural products of extremely widespread distribution. All of these compounds are mutagenic in bacteria and mammalian cells and MDA has been reported to be carcinogenic. MDA and analogs of its Beta-hydroxy-acrolein tautomer display an unusual pattern of mutagenicity in Salmonella typhimurium relative to simple acroleins and Alpha-dicarbonyl compounds. We believe this is due to a unique cyclic, nucleic acid adduct that is formed by MDA and Beta-substituted-acroleins. We propose to identify and quantitate the adducts formed on reaction of MDA and Beta-substituted-acroleins with DNA and compare the adduct levels to the mutagenicity of each of the compounds. We will also compare the pattern of genetic changes induced in frameshift and base-pair substitution strains of Salmonella. We will investigate the structure of MDA adducts to DNA and to oligonucleotides by optical and nuclear magnetic resonance spectroscopy and compare the genetic consequences of the incorporation of each adduct into a defined sequence of viral DNA. MDA, acroleins, and dicarbonyl compounds are an important new class of potential carcinogens. We will determine their tumor-initiating activity in Sencar mouse skin. MDA has been reported to exist in large quantities in tissue and plasma based on colorimetric assays. These reports have led to speculation that agents that elevate MDA levels enhance the risk of developing cancer (e.g., copper-containing IUD's for the uterine cervix). We propose to use an HPLC method that we have developed to accurately quantitate MDA levels in tissue and plasma. These studies will provide a quantitative basis with which to evaluate the carcinogenic potential of MDA and structurally related compounds. They will also generate important information about the mechanism of induction of mutation by dicarbonyl and Alpha, Beta-unsaturated carbonyl compounds, an extremely widely distributed group of natural and synthetic chemicals.